Hou et al. compared the efficacies of telbivudine and lamivudine in patients with hepatitis B e antigen (HBeAg)-positive and HBeAg-negative chronic hepatitis B. 1 Virological response definition in the study (Ͻ5 log 10 copies/mL) can not be applied to HBeAg-negative patients. HBeAg-negative patient usually harbor lower hepatitis B virus (HBV) DNA levels. 2 Also in the study, median HBV DNA levels of HBeAg-positive patients are more than 100-fold (Ͼ2 log 10 copies/mL) higher than those of HBeAgnegative patients. Defining the virological response as Ͻ5 log 10 copies/mL in both HBeAg-negative and HBeAg-positive patients might overestimate the response rate in HBeAg-negative patients. Accordingly, the authors reported the therapeutic response of telbivudine as 100% whereas the polymerase chain reaction (PCR)-negative HBV DNA rate was 85% in this small study population.
What should be the virological response definition in HBeAgnegative patients? A recent review defined it as "decrease of serum HBV DNA to PCR-undetectable levels (preferably) or Ͻ2000 IU/mL (4 log 10 copies/mL)." 3 The entecavir study used "HBV DNA Ͻ300 copies/mL by PCR assay" 4 and the adefovir study used "Ͻ400 copies/mL". 5 A recent clevudine study, finally, used hybride capture with a lower limit of detection of Ͻ4700 copies/mL and then, if it remains undetectable, used PCR with a lower limit of detection of 300 copies/mL. 6 Instead of an arbitrary limit, HBV DNA levels should have some clinical or virological predictive roles. For example, in an adefovir study, decrease of HBV DNA levels to Ͻ 10 3 copies/mL at week 48 of therapy has been found to be highly predictive of maintenance of the response at year 3 with a probability of HBV resistance as low as Ͻ 4%. 2,7 Clinical trials of HBV treatment in the current literature are not easy to compare directly. Patient demographics, baseline viral load, genotypes, and many other features already differ. However, at leastinclusion criteria and response definitions should allow making simple comparisons.